Hyland.DocumentFilters 25.3.0

• .NETFramework 3.5

  • No dependencies.

• .NETStandard 2.0

  • No dependencies.

# Enhancements

- Introduced support for HL7 Medical Records, including detection and processing into both Markdown and HiDef output modes. This enables HL7 content to be transformed into structured, machine-readable formats for use in AI workflows, data integration, and analysis while maintaining fidelity to the original record structure. (DF-2467)
- Introduced support for OCR processing of embedded images within document pages when exporting to HiDef output. The new `OCR_INLINE_IMAGES` option detects and extracts text from inline images, ensuring that more context is preserved for downstream uses such as AI workflows, data extraction, and content transformation. Supported in this release for email and web archive formats (HTML, MSG, EML, MHT, ICS/VCAL, OLK15) and office productivity formats (Apple Keynote, Apple Numbers, Apple Pages, HWPX, OpenOffice, RTF, MS Word, WordPerfect, MS Visio, MS PowerPoint). This capability helps retain meaningful text content from screenshots, diagrams, and other inline images, ensuring it is available in structured outputs like Markdown without altering the original appearance in image-based formats. (DF-2384)
- Introduced support for file type identification of JSON files. (DF-2457)
- Introduced support for table detection in PDFs to better recognize tables that contain only inner borders without explicit outer borders. This enhancement ensures more accurate identification of grid-based data for downstream uses such as AI systems, structured data extraction, and document transformation workflows. (DF-2374)

# Updates

- HD Mode: Resolved a condition for Apple iWork files where images may rendered with low resolution. (DF-2503)
- HD Mode: Resolved a condition for Apple iWork files where padding may be incorrectly applied to the top of a textbox. (DF-2503)
- HD Mode: Resolved a condition for Apple iWork files where paragraphs may be rendered with a black background. (DF-2503)
- HD Mode: Resolved a condition for DOCX files where floating tables with disabled overlap were overlapping. (DF-2479)
- HD Mode: Resolved a condition for DOCX files where numbered lists were sometimes skipping sequence numbers. (DF-2491)
- HD Mode: Resolved a condition for DOCX files where the layout of inline images could lead to a Page Limit error. (DF-2222)
- HD Mode: Resolved a condition for DOCX where Unicode characters were not rendered properly in non-standard files. UTF-8 is now a default encoding for DOCX internal XML files. Added MSWORD_DEFAULT_UTF8 option allowing for a revert to previous behavior. (DF-2475)
- HD Mode: Resolved a condition for EML and MSG files where HTML elements with style "display:none" were excluded from output.  A new option "HTML_SHOW_DISPLAY_NONE" (defaulting to OFF) can be enabled to force display of hidden elements of HTML structures for EML and MSG files, which mimics the behavior of some email clients, like Outlook. (DF-2421)
- HD Mode: Resolved a condition where Microsoft Word documents created with Microsoft Office 2013 or later could have graphics placed in the wrong position. (DF-2248)
- HD Mode: Resolved a condition where table rows containing graphics and/or text boxes were not sized properly, and text boxes were positioned incorrectly with regards to graphic elements. (DF-2224)
- HD Mode: Resolved an issue where HTML-based files using indirect color values could render as black. (DF-2379)
- Resolved a condition for MSG files where files may render differently between Windows and Linux. (DF-2522)
- Resolved a condition for PDF files where punctuation at the end of the line may appear in the wrong location in RTL documents. (DF-2482)
- Resolved a condition for XLS files where files with an invalid SupBook could fail to render. (DF-2512)
- Resolved a condition for the OCR image callback where the SourcePageIndex may be incorrectly set when processing PDF files. (DF-2511)
- Resolved a condition of OCR callback hooks where a SIGSEGV may be raised when processing images <= 8 bits per pixel. (DF-2547)
- Resolved a condition where DOCX metadata output used "AppVersion" for the application version, while other Office formats used "App Version". DOCX now uses "App Version" for consistency in metadata naming across Office formats. (DF-2248)
- Resolved a condition where converting a PDF file with the only root reference being inside an xref object could cause a segmentation fault. (DF-2476)
- Resolved a condition where processing a corrupted PDF file with a xref stream could result in the file being recognized as unreadable instead of corrupted. (DF-2505)
- Resolved a condition where text returned from the Python bindings method Extractor.GetText(stripControlCodes=True) would be missing commas. (DF-2488)
- Security: 7-zip: CVE-2025-55188: Document Filters is not affected because the vulnerable code path is never invoked. The product requires callers to provide an explicit, validated extraction path and does not use p7zip to directly create files on the system. These runtime protections prevent exploitation of the symbolic link handling issue. (DF-2564)
- Security: Protobuf: CVE-2025-4565: not-affected: affected code is not utilized by Document Filters. (DF-2498)
- Security: international_components_for_unicode: CVE-2025-54627: code-not-present: issue is specific to genrb CLI tool build process, not in the ICU library itself. Document Filters does not use the genrb binary or the affected parsing code during runtime operation. (DF-2557)
- Security: libtiff: CVE-2024-13978: not-affected: affected code is not utilized by Document Filters.  (DF-2546)
- Security: libtiff: CVE-2025-8176: not-affected: affected code is not utilized by Document Filters.  (DF-2540)
- Security: libtiff: CVE-2025-8176: not-affected: affected code is not utilized by Document Filters. (DF-2558)
- Security: libtiff: CVE-2025-8177: not-affected: affected code is not utilized by Document Filters.  (DF-2539)
- Security: libtiff: CVE-2025-8534: not-affected: affected code is not utilized by Document Filters. (DF-2550)
- Security: libtiff: CVE-2025-8961: Document Filters vendors libtiff 4.6.0 and does not compile or ship the tiffcrop CLI utility (tools/tiffcrop.c not part of build; no tiffcrop objects found in build output). CVE only impacts tiffcrop in 4.7.0 per NVD/VulDB advisory; our version predates affected release and excludes the tool, making the vulnerable code unreachable. (DF-2565)
- Security: openjpeg: CVE-2025-50952: patched: Applied security patch to address the issue.  (DF-2555)
- Security: openjpeg: CVE-2025-54874: patched: Applied security patch to address the issue.  (DF-2552)
- Security: protobuf: CVE-2025-53605: not-affected: Vulnerability only affects rust protobuf code, those parts are not incorporated into Document Filters. (DF-2509)
- Security: skia: CVE-2025-54627: code-not-present: issue is specific to Huawei HarmonyOS’s implementation of Skia and its associated calling code, therefore Document Filters is not affected by this issue.  (DF-2554)
- Security: zlib: CVE-2025-4638: not-affected: affected code is not utilized by Document Filters. (DF-2465)
- Security: zlib: CVE-2025-4638: not-exploitable: the issue is specific to the PointCloudLibrary, which uses a modified version of the ZLib library. (DF-2463)
- Security: zlib: CVE-2025-5087: false-positive: the issue is specific to the Kaleris NAVIS systems, not a vulnerability of the library itself. (DF-2507)
- Security: zlib: CVE-2025-52937: not-affected: issue is specific to the PointCloudLibrary, which uses a modified version of the ZLib library. This issue does not affect Document Filters. (DF-2506)